From Query to Counsel: Structured Reasoning with a Multi-Agent Framework and Dataset for Legal Consultation
Mingfei Lu, Yi Zhang, Mengjia Wu, Yue Feng
Feedback
Why It Matters
The development of JurisCQAD and the JurisMA framework addresses critical challenges in legal AI, such as data scarcity and complex reasoning, paving the way for more effective legal consultation tools.
Contributions
- Creation of a large-scale, expert-annotated dataset for legal queries and a novel multi-agent framework for structured reasoning.
Insights
- The integration of modular agents allows for dynamic adaptation to complex legal queries.
Limitations
- The dataset is focused on Chinese legal queries, which may limit applicability to other legal systems.
Tags
- agent
- data
- llm
- reasoning
Abstract
arXiv:2604.10470v1 Announce Type: cross Abstract: Legal consultation question answering (Legal CQA) presents unique challenges compared to traditional legal QA tasks, including the scarcity of high-quality training data, complex task composition, and strong contextual dependencies. To address these, we construct JurisCQAD, a large-scale dataset of over 43,000 real-world Chinese legal queries annotated with expert-validated positive and negative responses, and design a structured task decomposition that converts each query into a legal element graph integrating entities, events, intents, and legal issues. We further propose JurisMA, a modular multi-agent framework supporting dynamic routing, statutory grounding, and stylistic optimization. Combined with the element graph, the framework enables strong context-aware reasoning, effectively capturing dependencies across legal facts, norms, and procedural logic. Trained on JurisCQAD and evaluated on a refined LawBench, our system significantly outperforms both general-purpose and legal-domain LLMs across multiple lexical and semantic metrics, demonstrating the benefits of interpretable decomposition and modular collaboration in Legal CQA.